3.895 analysis of argument? basic quantative flaw? (69)

Willard McCarty (MCCARTY@vm.epas.utoronto.ca)
Wed, 3 Jan 90 20:11:06 EST

Humanist Discussion Group, Vol. 3, No. 895. Wednesday, 3 Jan 1990.


(1) Date: Wed, 03 Jan 90 16:07:45 CST (26 lines)
From: "Bill Ball" <C476721@UMCVMB>
Subject: computer analysis of argument

(2) Date: 3 January 1989 (23 lines)
From: Willard McCarty <MCCARTY@vm.epas.utoronto.ca>
Subject: the quantative study of literature

(1) --------------------------------------------------------------------
Date: Wed, 03 Jan 90 16:07:45 CST
From: "Bill Ball" <C476721@UMCVMB>
Subject: computer analysis of argument

I am beginning a dissertation project on methodologies to make use of
empirical evidence in normative discourse. More specifically, I am
working within the Liberal paradigm and the discipline of political
science--with the ultimate aim of being able to say something useful
about the moral questions raised in public policy decision-making. I
am currently exploring the applied philosophy/political theory
literature relevant to this project: Dewey, Baier, Taylor, Toulmin, and
others. However I would like to begin developing some computer-based
tools to assist me. In general I have in mind something very like an
expert system for the analysis of argumentation. I have been tooling
around with Turbo Prolog on the PC and a bit with HyperCard on the Mac.
At this point I am extremely interested in finding out who else may be
working on computer analysis of argumentation and what applications
might be appropriate for my interests. I would greatly appreciate any
leads and also any contacts with individuals interested in helping me
focus my ideas (I have a draft prospectus which I can Bitnet).

Bill Ball Bitnet: c476721@UMCVMB
Dept. Political Science Internet: c476721@UMCVMB.MISSOURI.EDU
U of Missouri-Columbia


(2) --------------------------------------------------------------------
Date: 3 January 1989
From: Willard McCarty <MCCARTY@vm.epas.utoronto.ca>
Subject: the quantative study of literature

Let me draw your attention to an article in Computers and the Humanities
23 (1989) by Willie van Peer (Utrecht), `Quantative Studies of
Literature. A Critique and an Outlook', pp. 301-7.

In this article, Professor van Peer argues that quantative literary
analysis, though based on a kernel of truth, suffers from a fundamental
deficit because it views a literary work as a collection of isolated
objects rather than a context-bound process. It works well, he points
out, at lower levels of linguistic organization (e.g., for grammar and
lexis) but is by nature very poor for figurative and hence literary
language. He holds that quantative methods should rather be applied to
the reactions of real readers (i.e., with respect to reader-response
theory) and, in general, be related to the issues of literary theory.
He notes that these methods can in turn bring some rigour to theoretical
studies.

Comments -- after reading the article?

Yours, Willard McCarty